© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 41-GEMMASCOPE-2-RES-65K
    7. 14892
    Prev
    Next
    INDEX
    Explanations

    say "whether" or "if"

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_41_width_65k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    どのように
    0.90
     how
    0.86
     کیسے
    0.85
     bagaimana
    0.85
     얼마나
    0.80
     کہاں
    0.80
     ماذا
    0.79
     কিভাবে
    0.78
    是怎么
    0.78
    什么时候
    0.76
    POSITIVE LOGITS
    whe
    1.57
     whe
    1.55
     Whe
    1.40
    Whe
    1.36
     regardless
    1.30
    regardless
    1.27
     Regardless
    1.27
    Regardless
    1.24
     WHE
    1.24
     irrespective
    1.19
    Activations Density 0.026%

    No Known Activations